1. Field of the Invention
This invention relates to residential and commercial electric hot water heaters of capacity up to thirty gallons and more specifically to a method for automatically determining when to turn on/off the electric current to such heaters so as to reduce the expenditure for electricity without compromising the users requirement for hot water.
2. Description of the Related Art
An electric hot water heater is illustrated in FIG. 1 and is typically provided with a thermostatic control (1) and a heating element (2) which heats water in a contained vessel (usually a cylindrical tank)(3). An electric current (4) is allowed to flow through the heating element by the thermostatic control which is usually operated by a bi-metallic contact which closes when the temperature of the surrounding water (5) is below the thermostat's "set" temperature and by opening when the temperature of the surrounding water is above the thermostats "set" temperature. Generally, such controls are subject to hysteresis which allows the water temperature to deviate from the set value by .+-.5.degree. F. or .+-.10.degree. F. Thermostat "set" temperatures are typically between 120.degree. F. and 180.degree. F. The modem water heater is often fitted with "heat-trap" devices (6) (commonly, an antiflow valve fated to a low heat conducting plastic pipe) which minimize heat losses to the adjoining inlet (7) and exit (8) pipes due to natural convection and conduction of hot water.
With most electric hot water heater controls there is normally no attempt to regulate the flow of electric current to the heater other than by the thermostat, and the water contained within the heater remains at or near the "set" temperature 24 hours a day regardless of the use of zero, small, or large volumes of water. This results in significant energy losses to the environment during extended periods of non-use of hot water, which would otherwise be significantly reduced were the heater in the switched off condition. The heat loss to the environment (Btu loss) related to the difference between the tank water temperature and the temperature of the ambient is linear; other factors, such as the ambient air velocity, pipe losses, etc., are not linearly related but also affect the heat loss. The tank temperature, however, responds to these losses as an exponential mathematical function of time.
During periods of no hot water use, the typical heater control responds to the various environmental factors which cool the tank and hence the water therein, by cycling the electric current so as to maintain the water at the "set" temperature. It is noteworthy that when tank water temperatures are reduced, there are savings in energy that accrue over time due to the difference in tank to ambient temperature; however, the water that has cooled during this period must ultimately be returned to the normal elevated operating temperature and therefore this portion of energy that must be restored is not saved.
Therefore, in order to create significant energy savings, the tank water temperature should be reduced to a value close to the ambient temperature for a time long enough to render the energy required to return the water to the "set" temperature relatively small when compared to the savings resulting from eliminating differences between the tank water temperature and the ambient. Paradoxically, a poorly insulated tank without the benefit of "heat traps", for example, will cool more quickly and will therefore produce a greater savings in response to being turned off over a fixed period of time than a well insulated tank. Likewise, since a smaller volume of water will cool more quickly to the temperature of the environment, a smaller tank capacity will also produce a larger percentage savings. Also, a method which allows the stored hot water in the tank to be utilized (thus quickly lowering the tank temperature), remain unheated for a prolonged period and then reheated just prior to need, will produce the greatest savings over time.
The following patents disclose prior efforts to reduce the energy consumption of electric hot water heaters.
Bartok, et. al. (U.S. Pat. No. 4,535,931) discloses a pair of thermistors (or other mechanical means) as sensors located in intimate contact with the inlet and outlet pipes of a water heater, operating through an operational amplifier to sense the difference in temperature of these points to determine a two valued set point for the heater thermostat; "high" or "low". The intent is to reduce the tank temperature during periods of no hot water use. Bartok suggests insulating the thermistors from the ambient so that the pipe temperature is less affected by the surrounding space.
This method, although possibly providing some energy savings, suffers from an important drawback in that this control has no a priori knowledge of the present or future need for hot water. Therefore the tank temperature will cool considerably before the first use of hot water after a long period of non use; thus providing "colder" rather than "hotter" water to the first user. It is noteworthy that if the first user experiences a sufficient volume of available hot water after a long period of non use then the control has not saved an appreciable amount of energy. This will most often be the case since the heater is not turned off but rather only set back to a lower temperature.
A further disadvantage of the Bartok method is the inability of the control to predict non use of hot water and thus to allow the using up of hot water from the tank during periods just prior to a prolonged period of non use without triggering the "raise the thermostat set point condition". This would insure the water temperature is hottest after the last user and this would limit the ability of Bartok's control to lower the tank temperature quickly so as to gamer the most energy savings during the non use period, i.e. the most energy savings accrue when the tank temperature is close to the ambient temperature.
Further, since the specific heat of water is high, the signal to raise the thermostat to the high operating set point may reset (due to simultaneous cooling of the relatively small volume of water in the exit pipe and the heating-caused by conduction and back flow of the small volume of water in the inlet pipe) before the large volume of water in the tank reaches the high operating temperature and thus the tank temperature may remain perpetually warm rather than hot.
As most modem water heaters are fitted with heat traps to prevent heat losses from conduction and convection of heated water from both the exit and inlet pipes, a small volume of water use will raise the thermostat set point for a long period; this resulting from the inability of the inlet and exit pipes to approach each other's temperatures, especially when well insulated from the environment (as Bartok suggests) and may render the control inoperative.
Another disadvantage of this scheme occurs when the ambient in the space (surrounding the water heater) is heated during the day, and left unheated at night. Under these conditions the water temperature in the inlet pipe could vary by 35.degree. F. making the setting of the threshold for raising and lowering the thermostat "set" point and its associated hysteresis impractical to determine. Also, the heater inlet pipe may be subject to a long pipe run within a heated space, such as might occur in an office complex or industrial setting, thus rendering the inlet temperature constant and since the functioning of the control makes use of the dropping inlet water temperature relative to the exit water temperature, making it difficult to find the correct threshold temperature difference to switch to the high temperature setting; as a result the control may not function as intended.
Kenneth W. Scott discloses an invention in U.S. Pat. No. 4,413,775 that is similar to the Bartok et al. patent and on which Bartok has improved. Scott employs two temperature sensing switches rather than thermistors, each on the inlet and exit pipes of the hot water heater and uses the set points of these switches to detect the 135.degree. F. fixed high temperature set point on the exit pipe and a fixed 100.degree. F. on the inlet pipe. Scott states "During periods of low use, heat energy is wasted if the temperature of the water remains at a high level. For example, a temperature of 120.degree. F. is adequate for all ordinary purposes, and a higher temperature does not ordinarily result in a higher temperature of the water actually used, but merely means that the user will mix more cold water with the hot for showering." Scott fails to mention that a portion of the energy not supplied to the system while the water cools exponentially from the high 135.degree. F. setting to the lower 120.degree. F. setting is not actually saved but must be returned to the system when again raising the water temperature to the high setting; while the energy saved is related linearly to the average tank temperature difference between the high and low setting. In order to effect significant savings, the tank temperature must be allowed to drop significantly from the high temperature set point and remain at that level for a long period of time.
Scott has also made some improvements to U.S. Pat. No. 4,166,944 which climatizes the inlet water temperature sensor threshold to 5.degree. F. to 10.degree. F. above the seasonal inlet water temperature. However, a heat trap on the inlet pipe (commonly installed on modem water heaters) will prevent the inlet pipe to refill with heated water from the tank and the control will not operate as implied.